The Lorenz system is one of the most famous examples in nonlinear dynamics and chaos theory. It is defined by the three coupled first-order equations
\[
x'=\sigma(y-x),\qquad
y'=x(\rho-z)-y,\qquad
z'=xy-\beta z.
\]
Although these equations are deterministic, their solutions can behave in a way that is extremely sensitive to initial conditions. That is one of the main signatures of chaos.
Historically, the Lorenz system arose from simplified atmospheric convection modeling. The variables \(x\), \(y\), and \(z\) can be interpreted as idealized flow and temperature quantities, while the parameters \(\sigma\), \(\rho\), and \(\beta\) control how the system evolves. For certain parameter ranges, especially the classic choice
\[
\sigma=10,\qquad \rho=28,\qquad \beta=\frac{8}{3},
\]
the trajectory does not settle to a fixed point or a simple periodic orbit. Instead, it traces out the famous butterfly attractor.
A remarkable feature of the Lorenz attractor is that it remains bounded while never repeating exactly. The motion swings around one lobe of the butterfly, then unpredictably jumps to the other lobe, and continues in a complex pattern. This is not randomness from outside noise; it comes from the internal nonlinear dynamics of the system itself.
To simulate the Lorenz equations numerically, one uses a time-stepping method such as the fourth-order Runge–Kutta method (RK4). At each step, the current state
\[
(x_n,y_n,z_n)
\]
is updated using several intermediate slope evaluations. This gives a good approximation to the true trajectory when the time step is chosen small enough.
For the sample input
\[
(x_0,y_0,z_0)=(1,1,1),\qquad t_{\max}=30,
\]
with the classic parameters, the resulting path typically winds around the two lobes of the attractor. If a second initial point is chosen very close to the first, for example
\[
(x_0+\varepsilon,y_0,z_0),
\]
with \(\varepsilon\) very small, the two trajectories can stay close for a short time but then separate noticeably. That sensitive dependence on initial conditions is one of the defining ideas of chaos theory.
A useful way to visualize this is through planar projections such as the \(x\)-\(y\) or \(x\)-\(z\) projection. Although the full motion is three-dimensional, these projections already reveal the butterfly shape clearly. Another useful quantity is the separation
\[
d(t)=\sqrt{(x_1-x_2)^2+(y_1-y_2)^2+(z_1-z_2)^2},
\]
which measures how two nearby states diverge over time.
At more advanced level, one studies Lyapunov exponents, strange attractors, fractal geometry, and bifurcation theory. But even this teaser version shows the central lesson: a deterministic system with simple equations can produce highly intricate long-term behavior.
This calculator uses RK4 to integrate the Lorenz system, display attractor projections, and compare nearby initial conditions so you can visualize chaotic sensitivity directly.
